When the requested authentication bitmask includes NTLM, we cannot
re-use a connection for another username/password as we then risk
re-using NTLM (connection-based auth).

This has the unfortunate downside that if you include NTLM as a possible
auth, you cannot re-use connections for other usernames/passwords even
if NTLM doesn't end up the auth type used.

Reported-by: Paras S
Patched-by: Paras S
Bug: http://curl.haxx.se/mail/lib-2014-01/0046.html

A regression introduced in 7f3b87d8 (present in the 7.21.4 release)
broke the total parser. Now skip the whitespace and the digits.

Reported-by: Justin Maggard
Bug: http://curl.haxx.se/mail/lib-2014-01/0019.html

When the progress callback returned 1 at a very early state, the code
would not make CURLE_ABORTED_BY_CALLBACK get returned but the process
would still be interrupted. In the HTTP case, this would then cause a
CURLE_GOT_NOTHING to erroneously get returned instead.

Reported-by: Petr Novak
Bug: http://curl.haxx.se/bug/view.cgi?id=1318

This is a debug function only and serves no purpose in production code,
it only slows things down. I left the code #ifdef'ed for possible future
pipeline debugging.

Also, this was a global function without proper namespace usage.

Reported-by: He Qin
Bug: http://curl.haxx.se/bug/view.cgi?id=1320

If OpenSSL is built to support SSLv2 this brings back the ability to
explicitly select that as a protocol level.

Reported-by: Steve Holme
Bug: http://curl.haxx.se/mail/lib-2014-01/0013.html

Some feedback provided by byte_bucket on IRC pointed out that commit
db11750c wasn’t really correct because it allows for “upgrading” to a
newer protocol when it should be only allowing for SSLv3.

This change fixes that.

When SSLv3 connection is forced, don't allow SSL negotiations for newer
versions.  Feedback provided by byte_bucket in #curl.  This behavior is
also consistent with the other force flags like --tlsv1.1 which doesn't
allow for TLSv1.2 negotiation, etc

Feedback-by: byte_bucket
Bug: http://curl.haxx.se/bug/view.cgi?id=1319

Since ad34a2d5 (present in 7.34.0 release) forcing
SSLv3 will always return the error "curl: (35) Unsupported SSL protocol
version" Can be replicated with `curl -I -3 https://www.google.com/`.
This fix simply allows for v3 to be forced.

Following commit 0aafd77f, replaced the internal usage of
FORMAT_OFF_T and FORMAT_OFF_TU with the external versions that we
expect API programmers to use.

This negates the need for separate definitions which were subtly
different under different platforms/compilers.

Added support to the built-in printf() replacement functions, for these
non-ANSI extensions when compiling under Visual Studio, Borland, Watcom
and MinGW.

This fixes problems when generating libcurl source code that contains
curl_off_t variables.

warning: 'res' may be used uninitialized in this function

Fixes a bug when all addresses in the first family fail immediately, due
to "Network unreachable" for example, curl would hang and never try the
next address family.

Iterate through all address families when to trying establish the first
connection attempt.

Bug: http://curl.haxx.se/bug/view.cgi?id=1315
Reported-by: Michal Górny and Anthony G. Basile

sendf.c:450:81: warning: Longer than 79 columns

Introduced in commit 2a4ee0d2 sending of data via the FILE
protocol would always return CURLE_WRITE_ERROR regardless of whether
CURL_WRITEFUNC_PAUSE was returned from the callback function or not.

Make sure that we detect such attempts and return a proper error code
instead of silently handling this in problematic ways.

Updated the documentation to mention this limitation.

Bug: http://curl.haxx.se/bug/view.cgi?id=1286

Previously this memdebug free() replacement didn't properly work with a
NULL argument which has made us write code that avoids calling
free(NULL) - which causes some extra nuisance and unnecessary code.
Starting now, we should allow free(NULL) even when built with the
memdebug system enabled.

free(NULL) is permitted by POSIX

free() itself allows a NULL input but our memory debug system requires
Curl_safefree() to be used instead when a "legitimate" NULL may be freed. Like
in the code here.

Pointed-out-by: Steve Holme

... since pthread_t may be non-scalar and/or may represent a real thread
with scalar 0.

Bug: http://curl.haxx.se/bug/view.cgi?id=1314

If a user indicated they preferred to authenticate using a SASL
mechanism, but SASL authentication wasn't supported by the server, curl
would always fall back to clear text when CAPABILITY wasn't supported,
even though the user didn't want to use this.

If a user indicated they preferred to authenticate using APOP or a SASL
mechanism, but neither were supported by the server, curl would always
fall back to clear text when CAPA wasn't supported, even though the
user didn't want to use this.

This also fixes the auto build failure caused by commit 6f2d5f05.

Fixes commit 1deac31e

This commit replaces that of 9f260b5d because according to RFC-2449,
section 6, there is no APOP capability "...even though APOP is an
optional command in [POP3].  Clients discover server support of APOP by
the presence in the greeting banner of an initial challenge enclosed in
angle brackets."

The FILE:// code doesn't support this option - and it doesn't make sense
to support it as long as it works as it does since then it'd only block
even longer.

But: setting CURLOPT_MAX_RECV_SPEED_LARGE would make the transfer first
get done and then libcurl would wait until the average speed would get
low enough. This happened because the transfer happens completely in the
DO state for FILE:// but then it would still unconditionally continue in
to the PERFORM state where the speed check is made.

Starting now, the code will skip from DO_DONE to DONE immediately if no
socket is set to be recv()ed or send()ed to.

Bug: http://curl.haxx.se/bug/view.cgi?id=1312
Reported-by: Mohammad AlSaleh